ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

A Novel Approach for Secure Mining of Horizontally Distributed Databases?

Journal: International Journal of Computer Science and Mobile Computing - IJCSMC (Vol.3, No. 7)

Publication Date:

Authors : ; ;

Page : 602-606

Keywords : Association Rules; Fast Distributed Mining; Frequent Item-sets; Horizontally Distributed Databases; Map Reduce; Privacy;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

The current protocols for secure mining of association rules in horizontally distributed databases has different Problems like privacy. In this kind of scenario Different players may wish to protect & preserve their data and metadata about their information. The approach is based on the Fast Distributed Mining (FDM) algorithm which is devised by Cheung [5]. FDM is distributed version of the Apriori algorithm and unsecured too. Most importantly these protocols provides two fresh secure algorithms (sub protocols) ? the union of private subsets that each of the participating players hold is carried out by one and other one checks for the inclusion of an element held by one of the player in a subset which is held by another. Our systems provides optimized privacy with respect to the protocol in [7] using Symmetric Key Cryptography. In addition, relatively it is more efficient in terms of computational cost, communication rounds and also in communication cost because of MAP/REDUCE

Last modified: 2014-07-25 22:05:59